Evaluation Method of Autonomous Controllability of Code Generator Based on Fuzzy Information Entropy

Research on the autonomous controllability assessment of code generators is of great significance for improving the efficiency of code generation, ensuring the accuracy of generated code, enhancing code maintainability and readability, and promoting the development of code automation technology. A fuzzy information entropy-based method for assessing the autonomous controllability of code generators is proposed to address the issues of high calculation error rate in evaluation index weight computation, low evaluation accuracy, and long evaluation time in traditional autonomous controllability assessment methods. Assuming the code generator as a random event, a fuzzy information entropy computation model is constructed to calculate the weights of evaluation indicators for assessing the autonomous controllability of code generators. This enables the selection of evaluation indicators and the construction of an evaluation indicator system. The evaluation indicator weights are calculated using the C-OWA operator, and the autonomous controllability assessment model is built using the maximum deviation minimization method. Particle swarm optimization is applied to optimize the model, and the solution is obtained using the discretized sub-swarm algorithm to achieve the autonomous controllability assessment of code generators. Experimental results show that the maximum error rate of evaluation indicator weight calculation in this method is 4.81 %, the evaluation accuracy varies between 97% and 88%, and the maximum evaluation time is 0.95s. This method can achieve accurate assessment of the autonomous controllability of code generators.